Show HN: TenseMem – deterministic memory for AI agents

Category: ai-ml

Tags: ai-agents, memory, sqlite, llm, deterministic

Score: 6.8/10 (Innovation: 6, Technical: 7, Documentation: 8, Utility: 6)

TenseMem is an embedded, deterministic memory layer for AI agents that stores typed facts in a single SQLite file and resolves superseded values (e.g., a new city replacing an old one) so models read one current truth instead of competing similarity matches. It adds freshness signals, correction history, and self-echo fencing, with adapters for OpenAI, Anthropic, Ollama, and MCP.

Target audience: backend and AI application developers building agent memory systems

Repository: https://github.com/tensemem/tensemem · Python · Apache-2.0 · 4 stars

View on Hacker News